1 Introduction

GPS radio sources have small sizes (< 500 pc) high radio luminosity (> 10²⁶ W Hz^-1) and are characterized by a convex radio spectrum peaking around 1 GHz. The nature of these radio sources hosted by galaxies and quasars is still debated. They could be the early stages of powerful extended radio sources or they could be indefinitely confined to their small sizes by a high density medium which prevents the expansion of the jets and lobes outside the central region of the galaxy hosting the radio source. A review of the properties of these sources is given by O'Dea (1998).

Due to the small sizes of GPS radio sources, radio observations at the milliarcsecond scale with VLBI techniques are necessary to reveal their morphology. The radio morphology is a critical piece of information necessary to our understanding of the nature of these sources (e.g., Wilkinson et al. 1994). The sidedness - i.e., to what extent the sources are two sided (CSO) or one sided (core-jet) - constrains the effects of Doppler boosting and is related to the Lorentz factor and orientation of the jets. The morphology (i.e., Fanaroff and Riley class I or II) provides clues to the transport of energy in the jets, e.g., transonic and non-relativistic or supersonic and relativistic, respectively. The extent to which the two sides of the source are similar/symmetric also constrains the interaction of the jet with the environment, i.e., does the jet remain well collimated or it is disrupted through interaction with dense gas? The morphology also provides constraints on how the sources may evolve (e.g., Readhead et al. 1996a,b; O'Dea & Baum 1997). In order to achieve this goal we have undertaken a VLBI study to observe the objects of the GPS 1 Jy sample (Stanghellini et al. 1998) and several of the GPS candidates listed by O'Dea et al. (1991)

Stanghellini et al. (1997) present global VLBI 6 cm images of 9 radio sources of the GPS 1 Jy sample. Here we add 11 radio sources from the list of O'Dea et al. (1991). Two of these objects are also included in the complete sample. Most of the sources presented here have never been previously observed before at this frequency.

In following papers in this series, images at other frequencies and a complete discussion of the results of our study will be given.

H₀=100 km s^-1 Mpc^-1, and q₀=0.5 have been assumed throughout this paper.

  

Figure 1: UV coverage for 1543+005

  

Table 1: Antennas involved in the VLBI experiment. The nominal system equivalent flux density (SEFD) in Col. 3 is given by the ratio between the system temperature (in K) and the antenna gain (in K/Jy). The VLA was available as a phased array for a fraction of time (equivalent values in parentheses) and with a single antenna otherwise

  

Figure 2: UV coverage for 0039+230

  

Figure 3: UV coverage for 0646+600